Data processing using a neural network system

A method of processing data using a processor configured to execute a neural network system, the method comprising:

• receiving input data at a merged layer, shared between a first neural network and a second neural network of the neural network system, from a prior layer of at least one of the first and second neural networks, wherein the first neural network comprises at least one subsequent layer in the first neural network and wherein the second neural network comprises at least one subsequent layer in the second neural network, wherein the merged layer precedes the at least one subsequent layer of the first neural network and the at least one subsequent layer of the second neural network, and wherein the at least one subsequent layer of the first neural network is separate from the at least one subsequent layer of the second neural network;
• applying a superset of weights to the input data to generate intermediate feature data representative of at least one feature of the input data, the superset of weights being combined from a first set of weights associated with the first neural network and a second set of weights associated with the second neural network;
• and
• outputting the intermediate feature data to the at least one subsequent layer of the first neural network and to the at least one subsequent layer of the second neural network.

A processor for executing a neural network system, the processor configured to:

• receive, for a merged layer, input data, a first set of weights associated with a first neural network and a second set of weights associated with a second neural network, wherein the first neural network comprises at least one subsequent layer in the first neural network and wherein the second neural network comprises at least one subsequent layer in the second neural network, wherein the merged layer precedes the at least one subsequent layer of the first neural network and the at least one subsequent layer of the second neural network, and wherein the at least one subsequent layer of the first neural network is separate from the at least one subsequent layer of the second neural network;
• apply a superset of weights to the input data to generate intermediate feature data for the merged layer, the intermediate feature data representative of at least one feature of the input data, the superset of weights being combined from the first set of weights associated with the first neural network and the second set of weights associated with the second neural network;
• and
• output the intermediate feature data to the at least one subsequent layer of the first neural network and to the at least one subsequent layer of the second neural network.